Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A comparative study of different machine learning methods on microarray gene expression data.

Mehdi Pirooznia1, Jack Y Yang, Mary Qu Yang

  • 1Department of Biological Sciences, University of Southern Mississippi, Hattiesburg 39406, USA. mehdi.pirooznia@usm.edu

BMC Genomics
|April 17, 2008
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

On-Demand Submetabolome Profiling of Early Ferroptosis with Porous Polymeric Magnetic Chemoselective Probes.

Analytical chemistry·2025
Same author

Cagrilintide-Semaglutide in Adults with Overweight or Obesity.

The New England journal of medicine·2025
Same author

Immunotherapy against colorectal cancer via delivery of anti-PD-L1 nanobody mRNA.

eGastroenterology·2025
Same author

Knowledge, Perception, and Attitude Toward Voluntary Counseling and Testing for HIV Among Secondary School Students in Iringa Rural District: Descriptive Cross-Sectional Study.

JMIR public health and surveillance·2025
Same author

From paradise to precarity: Hawai'i's public health gaps foreshadow U.S. coastal vulnerabilities.

Lancet regional health. Americas·2025
Same author

Dual Therapeutic Impact of AXL Inhibitor AB-329: Chemotherapy Sensitization and Immune Microenvironment Reprogramming in TNBC.

International journal of molecular sciences·2025
Same journal

Chromosomal scale genome assembly of medicinal plant Sophora tonkinensis.

BMC genomics·2026
Same journal

Variant-specific RNA testing resolves variants of uncertain significance in exome testing.

BMC genomics·2026
Same journal

Kaiso overexpression promotes an interferon immune response in murine intestines.

BMC genomics·2026
Same journal

Genomic evidence of ecological flexibility and cross-niche CRISPR spacerome targeting phage-plasmid hybrids in Latilactobacillus curvatus.

BMC genomics·2026
Same journal

Fgf evolution in vertebrates: insights from cyclostomes.

BMC genomics·2026
Same journal

Metabolic reprogramming, oxidative stress, and mitophagy in JSRV Env-transformed BEAS-2B cells: insights from integrated transcriptomics and metabolomics.

BMC genomics·2026
See all related articles

This study compared gene expression analysis methods, finding that feature selection significantly impacts classification accuracy for microarray data. Integrating feature selection with classification improves gene identification.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Microarray analysis relies on gene expression data for biological insights.
  • Various classification and feature selection methods exist for identifying differentially expressed genes.
  • A comprehensive comparison of these methods for microarray data analysis is lacking.

Purpose of the Study:

  • To compare the efficiency of different classification, clustering, and feature selection methods for microarray gene expression data.
  • To evaluate the performance of these methods in class prediction and identifying significant genes.

Main Methods:

  • Applied and compared Support Vector Machine (SVM), Radial Basis Function (RBF) Neural Networks, Multilayer Perceptron (MLP) Neural Networks, Bayesian, Decision Tree, and Random Forest classifiers.

Related Experiment Videos

  • Utilized v-fold cross-validation for accuracy assessment.
  • Employed K-means, DBC, and Expectation-Maximization (EM) clustering algorithms.
  • Evaluated feature selection techniques including SVM-Recursive Feature Elimination (SVM-RFE), Chi Squared, and CSF.
  • Tested methods on eight distinct binary microarray datasets.

    • Main Results:

    • Classification, clustering, and feature selection methods showed varying efficiencies on the tested datasets.
    • The selection of feature selection methods, number of genes, and sample size substantially influenced classification success.
    • Error rates and accuracy varied significantly across different classification algorithms based on chosen features.
    • Feature selection is crucial for accurate classification of new samples.

    Conclusions:

    • The study provides a comparative analysis of common methods for microarray data analysis.
    • Results highlight the critical role of feature selection in improving classification accuracy.
    • An integrated approach combining feature selection and classification demonstrates capability in identifying significant genes effectively.